Optimization of Molecules via Deep Reinforcement Learning

2018/10/31 by Zhenpeng Zhou, Steven Kearnes, Li Li +2 · 2 citations
Computer Science · Mathematics · #cs.AI #cs.LG #stat.ML

paper · pdf · doi:10.1038/s41598-019-47148-x

arxiv created 2019/03/01 · arxiv updated 2020/06/22

Abstract

We present a framework, which we call Molecule Deep Q-Networks (MolDQN), for molecule optimization by combining domain knowledge of chemistry and state-of-the-art reinforcement learning techniques (double Q-learning and randomized value functions). We directly define modifications on molecules, thereby ensuring 100% chemical validity. Further, we operate without pre-training on any dataset to avoid possible bias from the choice of that set. Inspired by problems faced during medicinal chemistry lead optimization, we extend our model with multi-objective reinforcement learning, which maximizes drug-likeness while maintaining similarity to the original molecule. We further show the path through chemical space to achieve optimization for a molecule to understand how the model works.

Cited by

• Deciphering Scientific Reasoning Steps from Outcome Data for Molecule Optimization
• Molecular Lead Optimization via Agentic Tool Planning

Related

• Distributed Reinforcement Learning for Molecular Design: Antioxidant case
• Grammars and reinforcement learning for molecule optimization
• Multiple-objective Reinforcement Learning for Inverse Design and Identification
• Adopting Reinforcement Learning with Verifiable Rewards for Molecular Generation
• DrugR: Optimizing Molecular Drugs through LLM-based Explicit Reasoning